The immune system recognizes small protein differences between cancer cells and the normal tissue from which they derive. Immunotherapy (cancer vaccines and immune modulators) is an emerging modality that educates the immune system to recognize these differences. Most cancers are insidious in their initiation, and develop over 10 to 30 years making it difficult to know when best to vaccinate and for how long. With recent advances mapping the genetic and epigenetic changes in cancers, it is now possible to apply principles learned from virus-targeted vaccines to the development of immunotherapies targeting the pathways induced by early genetic changes (driver genes) that are critical to cancer initiation. Cancers develop through a long process involving interconnections between genetic, epigenetic, and inflammatory signaling pathways that are constantly changing within the target organ as cells transition from normal to pre-malignant, to invasive cancer. In pancreatic adenocarcinoma (PDAC), emerging evidence suggests that these early changes induce a tumor-tolerant (anti-inflammatory) immune environment that includes a desmoplastic fibroblastic component, as well as various types of immunosuppressive stromal cells that produce cytokines, which inhibit anticancer (pro-inflammatory) immune responses. In this application we will test the hypothesis that targeting mutated Kras, the first genetic alteration driving PDAC initiation, can reverse the progressive pro-carcinogenic tolerogenic polarization that is associated with premalignancy development and progression. Furthermore, we will test the mechanistic hypothesis that the suppressive cellular response is under epigenetic control, and that targeting mutated Kras together with agents that change the phenotypic polarity of one or more cell types responsible for this procarcinogenic progression, will lead to tumor regression. We will employ a mouse model genetically programmed to progress through three stages of premalignancy (pancreatic intraepithelial neoplasms (PanINs) of increasing grade) before developing invasive cancer. In Aim 1 we will identify and characterize the phenotype and function of procarcinogenic inflammatory cell populations that are initiated by mutated Kras before and after vaccination with Listeria monocytogenes (LM) bacteria genetically engineered to express mutated Kras (LM-Kras). In Aim 2 we will perform a genetic and epigenetic analysis of suppressive cell versus anti-PanIN immune populations isolated from mouse and human PanINs. In Aim 3 we will combine the LM-Kras vaccine with the DNA hypomethylating drug 5aza- dC (DAC) and test for a synergistic anti-tumor effect. This application has significant translational implications that will fulfill an important clinical need - primary prevention and ealy treatment of a deadly cancer. We expect that the findings will drive a proof of principle clinica trial in subjects followed in screening clinics who are at risk for developing PDAC.